278results about "Adaptive network" patented technology

An adaptive variable step-size LMS filter based on a hyperbolic tangent function and its filtering method and computer equipment belong to the field of filter technology, and solve the problem that the LMS algorithm improved based on a nonlinear function requires too many manual adjustments of parameters and that the convergence speed and convergence accuracy of the traditional LMS algorithm cannot be satisfied simultaneously. The method of the present invention includes: determining the filter tap length and the number of iterations; initializing the filter parameters; obtaining the filter input signal vector; determining the instantaneous expected signal of the filter; obtaining the instantaneous output signal; obtaining the instantaneous error signal of the filter and the accumulated amount of the error signal based on the instantaneous output signal; obtaining the instantaneous step size using a step-size update formula based on a hyperbolic tangent function; and continuously iterating the filter tap weight coefficient vector to construct an LMS filter using the filter tap weight coefficient vector obtained after the number of iterations. The present invention is suitable for improving the performance of the LMS filter, and the filter of the present invention can be widely used in various signal processing fields.

The invention relates to the technical field of narrow-band filters, and discloses an ultra-low power consumption narrow-band filter implementation method based on a digital phased array system, and the method comprises the following steps: S1, collecting input signal data, carrying out the adaptive preprocessing, and generating preprocessed signal data; s2, performing digital beam forming processing based on the preprocessed signal data to generate digital beam forming data; s3, performing narrowband filtering parameter matching processing according to the digital beam forming data to generate narrowband filtering parameter data; in the narrowband filtering processing process, disturbance of environmental electromagnetic interference to a target signal is identified and eliminated in real time through adaptive noise suppression and a multi-dimensional signal preprocessing mechanism, so that the system can guarantee the pointing precision of digital beam forming, the pointing offset problem in the space signal capturing process is avoided, the accuracy of narrowband signal extraction is improved, and the accuracy of narrowband signal extraction is improved. And the reliability of filtering output and the signal fidelity are enhanced.

The invention relates to the technical field of adaptive filters, and discloses a wide linear adaptive filter, which comprises the following steps of: performing tensor integral solution on a system weight vector at the moment of the wide linear adaptive filter to obtain two adaptive weight vectors; constructing an input matrix based on the sampling value of the input signal, and calculating an estimation signal of the moment based on the product of the input matrix of the moment and the two adaptive weight vectors; and calculating the difference value between the expected signal and the estimation signal at the moment, obtaining an estimation error signal, and updating the gain vector at the obtaining moment and the estimation error signal by using a minimum error mode criterion, updating the two adaptive weight vectors at the obtaining moment, and reconstructing the system weight vector at the moment of the wide linear adaptive filter. According to the invention, the recursive adaptive filtering algorithm resistant to impulse noise interference is realized with low calculation complexity by using weight decomposition and selection of a robust cost function.

The invention discloses a semi-supervised learning method based on a self-adaptive threshold value, belongs to the field of radio communication, and aims to extract general semantic features of signals through comparative learning so as to improve generalization of downstream modulation identification tasks. A self-adaptive threshold mechanism is introduced to generate a high-quality pseudo tag, and confirmation deviation is reduced through additional category information, so that the robustness of the model is improved; besides, a hierarchical encoder capable of learning Fourier filtering is designed to capture multi-scale semantic features of large-scale long-sequence signals, improve the calculation efficiency of a model and research and verify an algorithm based on a plurality of large-scale data sets, so that the accuracy of automatic modulation recognition is remarkably improved, and particularly, under the condition of less label data, the accuracy of automatic modulation recognition is greatly improved. Compared with a mainstream method, the method has higher generalization ability and robustness, and can effectively cope with signal diversity and noise interference in a real communication environment.

The invention relates to the technical field of adaptive filtering, discloses a variable-step robust adaptive filter and a filter network, and aims to solve the technical problem that a traditional adaptive filter cannot effectively consider convergence speed, steady-state precision and impulse noise interference resistance. A non-linear error factor is introduced, large errors caused by impulse noise are effectively suppressed through a generalized function form of the non-linear error factor, and the robustness of the filtering process is remarkably improved; meanwhile, a variable step length mechanism is provided, the theoretically optimal candidate step length is calculated on line by minimizing the mean square deviation of the next moment, and a self-adaptive target variable step length is generated through truncation and time smoothing processing, so that dynamic balance is realized between rapid convergence and low-steady-state maladjustment. According to the invention, the filter is expanded to the distributed network, each node follows a diffusion strategy of first updating and then combining, and global information sharing and performance collaborative optimization are realized by using a combined coefficient.

The invention discloses a steel rolling process-based bar temperature filtering control method, which relates to the technical field of steel rolling production and comprises the following steps of: acquiring bar temperature original data, corresponding steel rolling process parameters and steel grade component data in advance; and a material temperature feature library is constructed, steel grade components are identified in real time according to an online spectrum analysis technology, and a matched filtering parameter reference template is called. By constructing the material temperature feature library and combining the online spectrum analysis technology, accurate recognition of different steel types and intelligent calling of the matched filtering parameter template are achieved, the problem that a traditional filtering method cannot adapt to the characteristic difference of the steel types is effectively solved, the universality and pertinence of filtering control are remarkably improved, and the method is suitable for large-scale popularization and application. Effective smoothness of high-fluctuation temperature data is guaranteed, the problem of lag or excessive filtering caused by a fixed window is avoided, and the dynamic adaptability of the system to a complex steel rolling scene is greatly enhanced.

The invention discloses a self-adaptive filtering method and system for an intelligent electric energy meter, and relates to the technical field of power grid signal processing, and the method comprises the steps: obtaining a power grid voltage and current signal set collected by a target intelligent electric energy meter historically, and obtaining a signal noise spectrum data set and a noise level evaluation result through spectrum analysis and noise level evaluation; if the result exceeds a preset noise threshold value, identifying an interference type and building a noise adaptive filtering multi-channel, integrating and deploying the noise adaptive filtering multi-channel into a target intelligent electric meter to complete signal filtering test and effect evaluation, and dynamically optimizing and generating an electric energy meter adaptive filtering multi-channel based on the obtained multi-channel filtering effect index parameter; and finally, signal filtering closed-loop control is realized. The technical problem that a traditional electric energy meter is weak in filtering pertinence and cannot adapt to dynamic noise of a power grid in electric power related scenes with complex dynamic interference is solved, and the technical effects that the electric variable measurement accuracy of the electric energy meter is improved, and the adaptive capacity of the electric energy meter to different interference scenes is enhanced are achieved.

The invention provides a reduced Gaussian kernel adaptive filtering method and system based on nearest center estimation, and relates to the field of data processing, and the method comprises the steps: obtaining an initial parameter and a training set, and constructing an adaptive filtering system based on the initial parameter; performing cluster division on an input signal, obtaining a feature vector of the input signal by calculating a reduced Gaussian kernel, and selecting a corresponding sub-filter according to a cluster for prediction to obtain prediction output; constructing a target optimization problem according to a maximum correlation entropy criterion based on an error between prediction output and expected output, and converting the target optimization problem into a convex optimization problem; solving a convex optimization problem by hybridizing three conjugate gradients, and updating the weight of a sub-filter; and repeating the steps to converge the error of the adaptive filtering system, and completing the adaptive filtering process. According to the method, the problem of dimension disasters existing in a traditional feature mapping method is solved by reducing the Gaussian kernel, the calculation complexity is remarkably reduced, and meanwhile unnecessary calculation overhead is greatly reduced through nearest center estimation.

The invention discloses a sampling signal noise reduction method and device based on a wavelet basis function, equipment and a medium, and belongs to the field of signal processing, and the method comprises the steps: carrying out the preprocessing of an original sampling signal after the original sampling signal is obtained, and obtaining a processing signal; performing wavelet decomposition on the processing signal to obtain a wavelet coefficient, and correcting the wavelet coefficient to obtain a correction coefficient; and performing inverse wavelet transform on the correction coefficient to obtain a noise reduction signal. According to the invention, after the original sampling signal is obtained, the preset composite power quality disturbance signal is added to the original sampling signal, and the noise reduction processing is carried out according to the wavelet coefficient of the signal combined with the noise to obtain the noise reduction signal. According to the invention, by adding the noise, high-efficiency noise suppression can be realized while key disturbance characteristics are reserved, so that the precision of the signal after noise reduction is improved; as the signal features of the original signals are reserved, a reliable preprocessing scheme can be provided for high-precision data acquisition scenes such as new energy monitoring of the power grid according to the original features subsequently.

The invention relates to the technical field of anti-dry-burning liquid level monitoring, in particular to an anti-dry-burning liquid level monitoring control system based on a PCB capacitance sensing circuit, in the anti-dry-burning liquid level monitoring control system based on the PCB capacitance sensing circuit, a signal preprocessing module adopts self-adaptive filtering to dynamically suppress noise interference in an original capacitance signal, and a reliable preprocessing signal is generated; the boiling state recognition module decomposes the signal into a steady state component and a transient state component, judges the boiling state according to the matching degree of the transient state component and a boiling feature library, and outputs a freezing instruction to maintain operation when the matching degree is high, so that mistaken shutdown is avoided; meanwhile, in combination with the judgment that the steady-state component is continuously lower than the warning line, a trigger instruction is output when the matching degree is low, and the real dry burning risk is accurately captured; the intelligent execution module responds instantly according to the instruction, guarantees continuous operation of equipment during freezing, executes power cut-off and sound-light alarm during triggering, and eliminates the potential safety hazard of dry burning; according to the process, accurate distinguishing between boiling interference and real dry burning is achieved, and equipment safety and operation continuity are synchronously optimized.

The invention relates to the technical field of telephone communication, and discloses a robust recursive least square adaptive filter based on inverse QR decomposition and a storage medium, and the method comprises the steps: calculating an inner product of an adaptive weight vector at a moment and an input signal vector, and obtaining an output signal at the moment of the adaptive filter; calculating a difference value between the expected signal and the output signal at the moment, obtaining an estimation error signal, and constructing a nonlinear weighting function; recursively updating and acquiring an inverse correlation matrix of the moment input signal vector based on the moment input signal vector and transpose thereof, a nonlinear weighting function, the inverse correlation matrix of the input signal vector and the covariance of the estimated error signal, and constructing a block matrix containing the inverse correlation matrix of the moment input signal vector; inverse QR matrix decomposition and unitary rotation operation are sequentially carried out on the block matrix, an array before inverse QR decomposition and an array after inverse QR decomposition at the moment are obtained, a gain vector at the moment is constructed, and a self-adaptive weight vector at the moment is updated and obtained.

The invention relates to the technical field of electronic circuits, in particular to a filter circuit, a voltage filter control circuit, a chip and electronic equipment. The filter circuit comprises a counter, a first data selector, a digital operation circuit and a register, the counter is used for receiving the counting enable signal, outputting a counting mark value to the first data selector, determining a corresponding current logic operation mode according to the current counting mark value, and performing logic operation on the intermediate input parameter according to the current logic operation mode to obtain a first voltage error value after filtering processing; by adopting the filter circuit provided by the invention, required logical operation can be completed by virtue of the first data selector and the digital operation circuit, so that the first voltage error value is output, and the filtered voltage signal is adjusted according to the first voltage error value, so that the set frequency component in the output voltage signal is filtered out; in this way, complex filter coefficients do not need to be configured, a processor with higher computing power does not need to be used, and hardware cost is saved.

The invention provides a pulse signal adaptive filtering method and system based on multi-band signal entropy. The method comprises the following steps: receiving a plurality of satellite communication signals of different frequency bands, and extracting a phase offset, an amplitude value and a frequency value; and generating a multi-dimensional feature group of each frequency band, and extracting the signal-to-noise ratio and the frequency variation of the corresponding frequency band from the multi-dimensional feature group. And based on the signal-to-noise ratios and the frequency variations of all the frequency bands, dynamically adjusting filtering parameters through a preset three-stage tuning structure, and filtering to obtain a filtered signal. Transmitting the filtered signal to a federated learning framework, and calculating a multi-band signal entropy value by combining the multi-dimensional feature group; adaptive filtering parameters are optimized based on entropy values, global aggregation is carried out after differential privacy processing, and an aggregation feature graph group is generated; and through a channel and a space attention mechanism, effective information of a high-entropy frequency band is enhanced, noise of a low-entropy frequency band is suppressed, and pulse signal adaptive filtering based on multi-frequency-band signal entropy is completed. According to the technical scheme provided by the invention, the accuracy of adaptive filtering of the pulse signal is improved.

The invention relates to the technical field of radio frequency signal processing, and discloses a digital filtering repair method and system for signal distortion, and the method comprises the steps: obtaining a frequency domain amplitude-frequency response and a phase response; constructing a phase compensation amount by adopting piecewise polynomial fitting and a phase inversion mechanism; a complex number logarithm domain mapping mechanism is adopted to execute composite frequency response inversion; constructing a switchable hybrid digital filter with an amplitude path and a phase path coupled in parallel; multi-band adaptive scheduling is executed; and outputting the radio frequency baseband signal stream. The technical problems that in the prior art, frequency response nonlinear distortion in a broadband range cannot be dealt with by adopting a fixed template or a single-path filter, and particularly, a high-precision and low-delay dynamic compensation effect cannot be realized in a scene that multi-band coupling interference exists in a radio frequency ADC channel with the bandwidth exceeding 800MHz are solved. Due to the fact that the switchable hybrid digital filter is constructed, accurate modeling and flexible recovery of distortion characteristics in broadband signals are achieved.

The invention discloses a filtering adjustment method and system based on an adaptive clock, and relates to the technical field of filtering adjustment, and the method comprises the steps: carrying out the time domain framing processing of a collected input signal through employing a sliding window, and combining DCT and compressed sensing to reconstruct the signal, and the main frequency of the signal and the dynamic change of the main frequency in the time sequence are accurately extracted through FFT, Gaussian pyramid layering and a multi-scale optical flow method, so that the frequency identification capability of the non-stationary signal is effectively enhanced, the frequency spectrum resolution is ensured, the data processing amount is remarkably reduced, the response speed of the signal frequency change is improved, and the method is suitable for popularization and application. Through the main-sub clock structure, the quick response of the center frequency of the filter to the change of the main frequency is ensured, the selectivity of the filter and the signal fidelity are improved, through the compensation adjustment of environmental factors, the influence of temperature and voltage changes on signal processing can be effectively reduced, and the robustness and stability of the system are enhanced.

The invention discloses a method for detecting and inhibiting resonance of a two-axis photoelectric pod. Belongs to the technical field of photoelectric pod control systems, and particularly relates to the technical field of two-axis photoelectric pod resonance detection and suppression. According to the invention, a stable precision index is taken as a discrimination standard of a resonance condition, a self sensor is utilized, a resonance frequency automatic detection and identification method is adopted, filter parameters are adaptively adjusted according to resonance frequency point distribution characteristics, and the resonance frequency is suppressed through a signal processing algorithm, so that interference of the resonance frequency on a test is effectively avoided. According to the method, the self-adaptive capability of the photoelectric pod is enhanced, so that the photoelectric pod can cope with different external interference factors, and especially in a special vibration environment, the pod can keep relatively high stability and precision. According to the method, the problem of vibration interference in a traditional method can be effectively solved, and the adaptability of the photoelectric pod in a vibration environment can be remarkably improved.

The invention relates to the technical field of signal processing, and discloses an anti-interference electrically tunable wave trap operation control method and system, and the method comprises the steps: collecting an input signal of an adjustable wave trap group, and employing adaptive Kalman filtering to dynamically suppress strong electromagnetic noise to construct a reinforcement learning state vector, and generating an optimal action instruction of the tuning module by using the deep Q network reinforcement learning model so as to control the tuning module to perform parameter adjustment on the adjustable wave trap group, and finally updating and adjusting the deep Q network reinforcement learning model and the adaptive Kalman filtering based on an actual effect index. Therefore, the electrically tunable wave trap high-precision operation control method based on adaptive Kalman filtering and depth Q network depth collaboration is constructed, the electrically tunable wave trap can be helped to stably maintain high-precision wave trapping performance in a scene of complex dynamic interference and strong noise superposition, and the performance of the electrically tunable wave trap is improved. And many limitations of the traditional scheme in the aspects of interference adaptation, precision control, multi-frequency coordination and stability are thoroughly solved.

The invention discloses a voltage signal adaptive filtering method based on Logistic regression, which adaptively adjusts the window width according to the change of the standard deviation of a voltage signal, so that a signal segment can reflect the dynamic characteristics of the frequency spectrum in real time, thereby effectively avoiding the problem of detail loss or processing redundancy caused by the use of a fixed window. Besides, in order to solve the problem that high-frequency and low-frequency signals are difficult to process efficiently at the same time in a traditional method, a Logistic regression model is introduced to classify the windowed signals, and the most suitable filter is dynamically selected for processing according to a classification result, so that the filtering precision is remarkably improved, and the filtering efficiency is improved. And the adaptive capacity of the system in a complex and changeable multi-frequency signal environment is enhanced.

The invention relates to the technical field of servo control, in particular to a servo driving current harmonic compensation method and system for electromagnetic interference suppression. The method comprises the following steps: extracting common-mode and differential-mode interference by using current and voltage sampling and combining differential measurement, and configuring a multi-stage passive filter according to spectrum characteristics to construct a suppression channel; establishing a digital interference cancellation mechanism, generating a reverse compensation signal by adopting self-adaptive trapped wave and a DSP (Digital Signal Processor), and injecting the reverse compensation signal into a control loop to form a software-defined multi-channel suppression architecture; after interference suppression, a high-order tensor model is constructed for the purified three-phase current, and an alternating least square method is adopted for decomposition so as to separate multi-band harmonic waves; a graph neural network is introduced, harmonic coupling is represented, and dominant and residual harmonic decoupling is realized through message passing; according to a grading progressive strategy, dominant harmonic waves are firstly subjected to coarse compensation, then residual harmonic waves are finely adjusted, and a compensation sequence is adaptively optimized through iterative feedback, so that high-robustness closed-loop control of servo driving is realized.

The invention discloses a dynamic self-triggering filtering method under incomplete semi-Markov kernel information, relates to the technical field of automatic control and signal processing, and aims to solve the problems that full acquisition of semi-Markov kernel information is difficult, communication burden is heavy and peak control is insufficient. A self-triggering mechanism containing a dynamic variable is designed to determine a data transmission moment by classifying a known / unknown set of semi-Markov information and utilizing known parameters and an unknown lower bound, a modal-dependent filter is constructed, and the mean square index stability and robust performance of a filtering error system are verified in combination with the Lyapunov theory and a linear matrix inequality technology. The problems that an existing method is high in conservative property and wastes resources are solved.

The invention relates to the technical field of digital filters, and discloses a quantization sensing wide linear minimum error mode adaptive filter, which performs non-uniform low-bit quantization on an input signal and an expected signal by using a quantization parameter to obtain a corresponding quantization value; constructing a linear factor based on the quantization parameter; based on the non-roundness of the input signal, constructing a normalized correlation matrix of the input signal and the linearized error signal, and constructing a compensation matrix in combination with a linear factor; obtaining input signal quantized values at a current moment and a historical moment, constructing an augmented input signal vector, calculating a product of the augmented input signal vector and the compensation matrix, and constructing a corrected input vector; performing wide linear filtering on the corrected input vector to obtain an output signal, and calculating an error signal in combination with an expected signal quantized value; and updating the weight vector at the next moment by using the corrected input vector and the error signal. According to the invention, by introducing a non-uniform low-bit quantization and deviation compensation mechanism, the cost and energy consumption of the analog-to-digital converter are remarkably reduced, and a relatively low steady-state error is kept.

The invention relates to the technical field of underground pipeline detection, in particular to a Hertz signal digitization method and system for underground pipeline detection. A digital detection excitation signal with stable Hertz frequency is generated by adopting a phase accumulation mechanism of a DDS (direct digital synthesis) technology and digital interpolation translation; a controllable random jitter frequency division strategy is introduced to generate multiple groups of single-frequency-point voltage signals to reduce spectrum aliasing and background noise; a response signal is sampled and filtered through a high-precision ADC and a self-adaptive FIR low-pass filter; and carrying out detection analysis and alarm on the target underground pipeline in combination with the frequency band energy normalization analysis and feature vector construction of the wavelet sub-band. According to the invention, digital improvement can be carried out on a 512 Hz analog signal detected by a traditional underground pipeline, and the sensitivity, the anti-interference capability and the signal processing precision of underground pipeline detection are greatly improved.

The invention provides a learning enhanced Kalman filtering method for forest emergency positioning and a related device. The method comprises the following steps: acquiring time sequence measurement data of an unmanned aerial vehicle and a ground target node; inputting time sequence measurement data into an error compensation network, and outputting a ranging error compensation amount and an observation noise covariance matrix; performing distance measurement compensation on the time sequence distance measurement data, feeding back the distance measurement data subjected to error compensation and a state error of the ground target node to a state prediction network, and outputting a state transition matrix and a process noise covariance matrix; state prediction is carried out based on a state transition model and a process noise covariance matrix, Kalman filtering fusion is carried out on a prediction state, compensated distance measurement data and an observation noise covariance matrix, three-dimensional position estimation of a ground target node is determined, combined improvement of measurement correction and prior prediction precision is realized, and the accuracy of measurement correction and prior prediction is improved. On the basis of keeping the optimality of traditional filtering, modeling is carried out on time-varying noise and non-linear motion, and the positioning precision in a forest complex environment is improved.

The invention relates to the technical field of interference suppression, in particular to an electromagnetic compatibility optimization method and system for a ship shaft frequency converter, and the method comprises the steps: carrying out the term-by-term judgment of sensitive interference energy, current errors and switching loss after the discretization and jump limitation based on a frequency sequence, and screening out a boundary-crossing sequence, the genetic algorithm executes cross combination and amplitude superposition in the residual sequence, a new frequency set is formed through transformation and rearrangement, frequency points are distributed in a plurality of intervals to be reorganized, the switch state is predicted and updated one by one through Kalman filtering, observation residual errors are corrected in real time, and the frequency points are reorganized. The harmonic distortion rate, the radiation intensity and the instruction change amplitude are synchronously output by taking the time index as a sequence, the deviation is compensated at the amplitude and phase level, the harmonic index and the radiation index are controlled in an agreed range, the switching loss and the instruction change are within threshold values, and a complete closed loop covering frequency generation, harmonic correction and sequence updating is formed.

The invention discloses a robust adaptive fusion filtering astronomical attitude determination method and system, and belongs to the technical field of radio navigation. The method comprises the following steps: constructing an integrated navigation model of a starlight and inertial integrated navigation system based on Lie group description; obtaining an observation residual error and an observation residual error covariance of a corresponding mode; robust and adaptive mode parallel sub-filtering is carried out, and posterior states and covariances in the two modes are updated; updating the mode probability; and fusing the updated posterior state with the covariance to obtain fusion output of parallel sub-filtering, taking the fusion output of the parallel sub-filtering as final estimation of the starlight and inertial integrated navigation system based on Lie group description, and outputting an astronomical attitude determination result containing an azimuth angle, a pitch angle and a roll angle. The high-precision astronomical attitude determination method has strong adaptability, strong robustness and accuracy when coping with complex noise and measuring outliers, and can realize high-precision astronomical attitude determination of starlight and inertial integrated navigation in a complex dynamic environment.

The invention belongs to the technical field of battery residual life prediction, and particularly relates to a lithium ion battery residual life prediction method considering dynamic temperature influence, and the method specifically comprises the steps: building a lithium ion battery capacity degradation model, and obtaining a state space model under the dynamic temperature influence; a filtering method based on a Bayesian framework is used for predicting the capacity degradation state estimation value and the residual life of the lithium ion battery, a time-varying fading factor is introduced in the importance sampling process, and adaptive kernel density estimation is used in the resampling process. According to the method, a Wiener process is adopted to model a battery capacity degradation process in a nonlinear and time-varying temperature environment, and meanwhile, a time-varying fading factor and adaptive kernel density estimation are introduced to construct an importance density function and a resampling process of a filtering algorithm based on a Bayesian framework, so that the capacity degradation state change tracking capability of the filtering is improved, and the filtering efficiency is improved. And the battery residual life prediction error under the influence of the external environment is reduced.

The invention relates to the technical field of offshore wind power monitoring, in particular to an offshore wind power jacket foundation structure fatigue life intelligent monitoring system which comprises a signal acquisition module, a data preprocessing module, a fatigue feature extraction module, a fatigue damage calculation module, a life prediction module and a result output module. The signal acquisition module acquires a mechanical response signal of a jacket foundation structure and transmits the signal to the data preprocessing module, the data preprocessing module transmits the signal to the fatigue feature extraction module, the fatigue feature extraction module extracts feature parameters and transmits the feature parameters to the fatigue damage calculation module, and a fatigue damage value is calculated and transmitted to the life prediction module. The life prediction module predicts residual life and transmits a prediction result to the result output module, and the result output module outputs residual life data. According to the method, accurate monitoring and safe and economic operation and maintenance of the fatigue life of the jacket foundation are realized through arrangement of specific sensors, data processing, extraction of fusion features, damage calculation and model prediction and early warning.

The invention provides a segmented adaptive SG filtering method and device based on cosine weight transition. The method comprises the following steps: obtaining a noisy fault signal of a power distribution network; extracting local feature indexes of the signals, wherein the local feature indexes comprise an average gradient representing the change frequency of the signals and a composite feature index representing the noise level; dynamically determining a window length and a polynomial order of SG filtering based on the local feature indexes; the transient stage is the first 1 / 2 period of the post-fault signal, the steady-state stage is the subsequent part of the post-fault signal, an overlapping region is arranged at the junction of the transient stage and the steady-state stage, the overlapping region is a 1 / 2 period following the transient stage, and a cosine weight function is adopted to perform weighted fusion on filtering results of the two stages in the overlapping region; and splicing the filtering results of the non-overlapping region and the overlapping region of the transient stage and the non-overlapping region of the steady stage, and outputting a filtered fault feature signal.

The invention provides an LSTM-Kalman filtering method and device for unmanned aerial vehicle magnetic compensation, and belongs to the technical field of aeromagnetic signal processing, and the method comprises the steps: obtaining a reference geomagnetic field vector in a zero interference environment, and synchronously collecting an unmanned aerial vehicle attitude angle, OBE motor rotation speed, phase current, power supply noise and magnetometer data; oBE represents airborne electrons; removing a static magnetic field according to a reference value to obtain an effective magnetic disturbance variable, performing low-pass filtering, performing dynamic normalization on all features by using a sliding window, and constructing a time sequence feature matrix; inputting the time sequence characteristic matrix into a trained LSTM neural network, and outputting a magnetic interference predicted value; and calculating a residual error between the measured value and the predicted interference value to obtain a predicted true value, carrying out Kalman filtering smooth estimation on the predicted true value, and outputting a compensated geomagnetic field vector. According to the invention, integrity compensation of global magnetic disturbance is realized.

The invention provides a CMA-ES-based TI-DAC system ZOH characteristic frequency response compensation optimization method, belongs to the technical field of filter optimization, and aims to solve the problems that a traditional pre-compensation filter is designed for correction in a digital domain, the global search capability is weak, local extremum is likely to be trapped in, and the correction efficiency is low. The method comprises the following steps: S1, constructing a pre-compensation filter; S2, constructing a pre-compensation filter; s2, a filter is pre-compensated based on a CMA-ES method; s3, convolution is carried out on the input signal and a compensation filter, and compensation is carried out on the convolved digital signal in a digital domain through a pre-coding filter, so that sinc attenuation caused by ZOH is effectively compensated, and the problem of frequency response distortion in a TI-DAC system is solved; according to the method, the CMA-ES method is introduced, the coefficient of the FIR filter is optimized within the constraint frequency band range, and then convolution is carried out on the input signal and the compensation filter, so that sinc attenuation caused by ZOH is effectively compensated, and the problem of frequency response distortion in a TI-DAC system is solved.